236 Proceedings of Indiana Academy of Science 
“Sunken Island” (Fig. 1) consists of several acres at 4-10 feet below 
the surface, having a sand-marl bottom, and only small areas not 
covered with Potamogeton. Little evidence of movement of its abundant 
mussel population is ever seen. 
Mussels changed from one habitat to another usually exhibited 
greater unrest than undisturbed ones. A number were first accustomed 
to stream conditions, then subjected to the following experiment. They 
were placed, ten together in a rectangle, two siphoning upstream, two 
down, and the remaining six transversely to the current, in the mouths 
of Sugar and Pocahontas creeks. On succeeding days their positions 
were checked, with especial reference to tropic movements in response 
to current, depth, obstacles, distance moved, etc. The experiment was 
repeated many times. 
There were 101 identifiable reactions considered, as follows, in those 
cases in which some movement had occurred: 
(1) Remained transverse to current 25 out of 72 possible 
(2) Turned transversely to current 5 out of 48 possible 
(3) Remained faced downstream, siphoning up 23 out of 24 possible 
(4) Turned downstream, to siphon up 23 out of 96 possible 
(5) Remained facing upstream, siphoning down 13 out of 24 possible 
(6) Turned upstream to siphon down 12 out of 96 possible 
There was some tendency to remain transverse to the stream when 
placed that way originally, but much less tendency to assume a new 
position in opposition to current. There was a greater tendency to 
remain siphoning downstream than to turn that way. Of those set to 
siphon upstream nearly all retained that orientation, and one-fourth of 
the others assumed it, a much higher proportion than of those which 
chose to siphon in any other direction. This seems to bear out the 
tradition that mussels prefer to siphon upstream. Yet I am encouraged 
to believe that the orientation is as much a reaction to the pressure sense 
and a desire to reach deeper water, as it is a rheotropic reaction. 
Almost all the cases under item six took place after rains when the 
creeks rose; the depth of the water was doubled and the velocity in- 
creased. The reaction was more probably due to increased depth than 
to current. These cases are few but selected, and there were many 
rejected cases that seemed to point the same way. Yet an Anodonta 
placed in an eddy pool three feet deep did not move until a freshet 
raised the creek. Then it moved round and round the pool at the same 
depth, against the current of the eddy, not attempting to get out, and 
doubtless oriented by the eddy. 
The bottom of Sugar Creek consists of much finer sand, gravel, 
and mud than Pocahontas. Hence the movements of Lampsilis were 
much more frequent and pronounced in the former. This despite the 
fact that Sugar Creek is cleaner and colder. 
Movements were observed also in the outlet of the lake—Walnut 
Creek. When the dam was raised and the creek lowered most of its 
mussels sought deeper water, and more or less downstream movement 
took place. After periods of higher water the direction of movement 
was more random. Here again the amount of movement was coextensive 
with the favored bottom. 
